The accuracy of cystoscopy in predicting muscle invasion in newly diagnosed bladder cancer patients

PURPOSE: The prognosis of muscle-invasive bladder cancer (MIBC) has not improved for three decades. Transurethral resection of the bladder tumor (TURBT) is the standard procedure for local tumor staging. TURBT has several limitations, including the spread of tumor cells. Therefore, an alternative is needed in patients with suspected MIBC. Recent studies have shown that mpMRI is very accurate in staging bladder tumors. Because the diagnostic efficacy of urethrocystoscopy (UCS) has been reported as good as the efficacy of mpMRI to predict muscle invasion we performed this prospective multicenter study in which we compare UCS with pathology. METHODS: From July 2020 until March 2022, 321 patients with suspected primary BC in seven participating Dutch hospitals were included in this study. A flexible UCS was performed by urologists, physician assistants, or residents. Predictions of muscle invasion using a 5-point Likert scale alongside the histopathology data were recorded. The sensitivity, specificity, predictive values, and 95% confidence intervals were determined using a standard contingency table. RESULTS: Of the 321 included patients, 232 (72.3%) received a histopathological diagnosis of non-muscle-invasive bladder cancer (NMIBC) and 71 (22.1%) were histopathologically diagnosed as MIBC. In 2 patients (0.6%), classification was not possible (Tx). Cystoscopy predicted muscle invasion with a sensitivity of 71.8% (95% CI 59.9–81.9), and a specificity of 89.9% (95% CI 85.4–93.3). This corresponds to a positive predictive value (PPV) of 67.1% and a negative predictive value (NPV) of 91.7%. CONCLUSION: Our study shows a moderate accuracy of cystoscopy to predict muscle invasion. This result does not support the use of cystoscopy only instead of TURBT for local staging. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s00345-023-04428-6.